The critical role of SENP1-mediated GATA2 deSUMOylation in promoting endothelial activation in graft arteriosclerosis
Data from clinical research and our previous study have suggested the potential involvement of SENP1, the major protease of post-translational SUMOylation, in cardiovascular disorders. Here, we investigate the role of SENP1-mediated SUMOylation in graft arteriosclerosis (GA), the major cause of allograft failure. We observe an endothelial-specific induction of SENP1 and GATA2 in clinical graft rejection specimens that show endothelial activation-mediated vascular remodelling. In mouse aorta transplantation GA models, endothelial-specific SENP1 knockout grafts demonstrate limited neointima formation with attenuated leukocyte recruitment, resulting from diminished induction of adhesion molecules in the graft endothelium due to increased GATA2 SUMOylation. Mechanistically, inflammation-induced SENP1 promotes the deSUMOylation of GATA2 and IκBα in endothelial cells, resulting in increased GATA2 stability, promoter-binding capability and NF-κB activity, which leads to augmented endothelial activation and inflammation. Therefore, upon inflammation, endothelial SENP1-mediated SUMOylation drives GA by regulating the synergistic effect of GATA2 and NF-κB and consequent endothelial dysfunction.A potential role for SENP1 in vascular defects was suggested in our previous study31. To determine the expression pattern of SENP1 in clinical GA, similarly sized human coronary arteries from transplanted hearts with GA and from non-diseased hearts were collected for histological examination. SENP1 was weakly expressed in non-diseased vessels, whereas its expression was dramatically enhanced in the luminal endothelial layer (CD31+) of rejecting arteries; the changes were proportional to the severity of rejection (Fig. 1a,b). Elastic-Van Gieson staining showed obvious intimal expansion with abundant CD45+ inflammatory cell infiltration (Supplementary Fig. 1A–C). Thus, to further determine a link between endothelial SENP1 expression and vascular inflammation, VCAM-1 and P-selectin expression was assessed by immunofluorescence staining. As markers of EC activation, both VCAM-1 and P-selectin expression was markedly enhanced in the endothelium of the rejecting arteries compared with that of non-diseased controls, which is consistent with the pattern of SENP1 expression and inflammation within the vessel wall (Fig. 1c). Intriguingly, endothelial expression of GATA-2, a major transcription factor in EC activation, was elevated during rejection progression (Fig. 1d,e). These results suggest a positive role for SENP1 in EC activation and vascular inflammation during pathological vascular remodelling in GA.Tissue collection and analyses was approved by the Yale Human Investigation Committee and the New England Organ Bank. Human coronary arteries were collected with informed consent from cardiac transplant recipients with chronic rejection or healthy organ donors. Arteries were procured in the operating room, and the disease was diagnosed macroscopically by an experienced cardiac surgeon and coded without patient identifiers. Vascular samples were embedded in optimal cutting temperature compound and immediately frozen36.Supplementary Figures and Supplementary Table.This work was supported by the National Natural Science Foundation of China (81422005, 81270357, 81600354, and 31470057), the Zhejiang Provincial Natural Science Foundation of China (LR14H020002), the Fundamental Research Funds for the Central Universities of China, and the China Postdoctoral Science Foundation (2015M581924). This work was partly supported by National Key Research and Development Program of China (2016YFC1300600), National Natural Science Foundation of China (No. 91539110) to WM, R01 HL109420 and R01 HL115148, and CT Stem Cell Innovation Award (Established Investigator Grant) 14-SCB-YALE-17 to WM.The authors declare no competing financial interests.Author contributions L.Y. supervised the overall project. L.Y. and C.Q. designed the research. C.Q., Y.W., HG.Z., L.Q., Y.S., X.Z., L.S., X.Z., J.C., H.Z. and H.F.Z. performed the experiments and analysed the data. G.T. and W.M. provided critical comments. C.Q. and L.Y. wrote the manuscript.